The linear, steady-state, baroclinic model response to a tropical heating s
uperimposed on a three-dimensional basic state is examined in this study. T
he emphasis is on the relevance of the linear model solution as compared to
a fully nonlinear baroclinic model. The direct response to heating in the
fully nonlinear, time-dependent model is obtained as the day-30 model respo
nse, following the Jin and Hoskins approach. When a 15-day linear damping i
s included in addition to Rayleigh friction, Newtonian cooling, and a scale
-selective biharmonic diffusion, the comparison of the linear and the nonli
near model responses to a 2 degrees C/day tropical heating reveals a striki
ng similarity in both the spatial distribution and amplitude. Thus nonlinea
rity appears to be a secondary effect and may be crudely represented by the
15-day linear damping, and the linear steady-state model can be a useful t
ool in diagnostic studies.
Both the linear and the nonlinear model responses show an insensitivity to
heating longitudes, especially when heating is located between 30 degrees E
and 120 degrees W. This insensitivity is characterized by a geographically
fixed response that consists of a streamfunction center over the central N
orth Pacific and a weak wave train over the Pacific-North American region.
The spatial structure of the preferred pattern does not depend on the dissi
pation or the amplitude of the tropical heating in the nonlinear model. The
geographically fixed response is found to be prominent in the Northern Hem
isphere for both the northern winter and summer climatological basic states
.